System and method for distributing high-quality 3D video in a 2D format

Abstract

A checkerboard subsampled image is packed in a format to remove shifts that would otherwise result in jagged edges where edges that were not jagged used to exist when checkerboard subsampling is used to reduce video bandwidth. The packing format places alternate rows of a checkerboard subsampled image row by row or places alternate columns of a checkerboard image line by line. By removing the shift and jagged edges, high frequency components that are difficult to compress and that display as annoying artifacts are eliminated.

What is claimed is: 
     
       1. A system to provide 3D content to a viewer, comprising:
 a subsampler to subsample a stereoscopic 3D image; 
 a packer to pack the subsampled stereoscopic 3D image into a 2D compatible format having four quadrants; 
 a compression system to compress the packed data, wherein the compression system increases processing resolution at at least one quadrant boundary in the packed data; and 
 a transmitter to transmit the packed data to the viewer. 
 
     
     
       2. The system recited in  claim 1 , wherein the subsampler performs a checkerboard subsampling. 
     
     
       3. The system recited in  claim 1 , wherein the packer packs the data in a packing format that removes jagged edges that would otherwise cause artifacts when the data is compressed. 
     
     
       4. The system recited in  claim 2 , wherein the checkerboard subsampling results in a left eye subsampled image and a right eye subsampled image, and wherein the packer packs the data in a format such that alternate rows of the left eye subsampled image are placed row by row in the left eye image portion of the 2D compatible format, and alternate rows of the right eye subsampled image are placed row by row in the right eye image portion of the 2D compatible format. 
     
     
       5. The system recited in  claim 2 , wherein the checkerboard subsampling results in a left eye subsampled image and a right eye subsampled image, and wherein the packer packs the data in a format such that alternate columns of the left eye subsampled image are placed column by column in the left eye portion of the 2D compatible image, and alternate columns of the right eye subsampled image are placed column by column in the right eye portion of the 2D compatible image. 
     
     
       6. The system recited in  claim 1 , further comprising an encoder that performs higher resolution processing on the data at boundaries in the packed data to reduce undesirable noise in the compressed data. 
     
     
       7. A method for providing 3D content to a viewer, comprising:
 subsampling a stereoscopic 3D image; 
 packing the subsampled 3D image into a 2D compatible format that has four quadrants; 
 compressing the packed data using increased processing resolution at at least one quadrant boundary in the packed data; and 
 transmitting the packed content to the viewer. 
 
     
     
       8. The method recited in  claim 7 , wherein the subsampling comprises checkerboard subsampling. 
     
     
       9. The method recited in  claim 7 , further comprising packing the data in a packing format that remove jagged edges that would otherwise cause artifacts when the data is compressed. 
     
     
       10. The method recited in  claim 8 , wherein the checkerboard subsampling results in a left eye subsampled image and a right eye subsampled image, and wherein the packing comprises packing the data in a format such that alternate rows of the left eye subsampled image are placed row by row in the left eye image portion of the 2D compatible format, and alternate rows of the right eye subsampled image are placed row by row in the right eye image portion of the 2D compatible format. 
     
     
       11. The method recited in  claim 8 , wherein the checkerboard subsampling results in a left eye subsampled image and a right eye subsampled image, and wherein the packing packs the data in a format such alternate columns of the left eye subsampled image are placed column by column in the left eye portion of the 2D compatible image, and alternate columns of the right eye subsampled image are placed column by column in the right eye portion of the 2D compatible image. 
     
     
       12. The method in  claim 7 , further comprising an encoding the packed data at higher resolution processing at boundaries in the packed data to reduce undesirable noise in the compressed data. 
     
     
       13. A system to present 3D content to a viewer, comprising:
 a set top box to receive a stereoscopic 3D image in a 2D compatible format having four quadrants, and having no data shifts resulting from subsampling of the stereoscopic 3D image; 
 a decompression system to decompress the 2D compatible format data taking into account that the 2D compatible format data had been compressed using a increased quantization resolution at at least one quadrant boundary of the 2D compatible format data and a presentation system to display the decompressed 3D content to a viewer. 
 
     
     
       14. The system recited in  claim 13 , wherein the set top box repacks the received stereoscopic 3D image into a format compatible with the presentation system. 
     
     
       15. The system recited in  claim 13 , wherein the subsampling is checkerboard subsampling. 
     
     
       16. A method for presenting 3D content to a viewer, comprising:
 receiving a stereoscopic 3D image in a 2D compatible format, the 2D compatible format having four quadrants; 
 decompressing the received stereoscopic 3D image, wherein the decompressing takes into account that the 2D compatible data was compressed using increased quantization at at least one quadrant boundary; and 
 presenting the decompressed stereoscopic 3D image to a viewer. 
 
     
     
       17. The method recited in  claim 16 , further comprising repacking the received the stereoscopic 3D image into a format compatible with the presentation system. 
     
     
       18. The method recited in  claim 16 , wherein the subsampling is checkerboard subsampling.